Vol. 39, issue 04, article # 2

Abstract:

Expanding the instrumental and analytical methods for identifying harmful impurities in the atmosphere is an important task for solving environmental problems. In this regard, the work focuses on developing a comprehensive approach to detection of harmful impurities in atmospheric air. This approach is based on measurements of the absorption spectra of air containing harmful impurities along a path by pulse terahertz spectroscopy methods. To analyze the obtained spectral data, a neural network is created and applied, and arrays of model absorption spectra of gas mixtures with different qualitative and quantitative compositions are generated for its training. It is shown that the neural network is capable of identifying six gas components in concentrations of up to 0.01 ppm with accuracy of 90–95%. A series of experiments with real gases confirms the sensitivity of the THz spectroscopy method to low gas concentrations in the mixture. The results show that the combined method is sufficiently sensitive for identifying both single gases and gas mixtures, which can be used for environmental monitoring.

Keywords:

terahertz spectroscopy, neural network, deep convolutional neural networks, 1D convolutional networks, transformation of neural network architecture, gas analysis

Figures:

References:

1. Bassous N.J., Rodriguez A.C., Leal C.I.L., Jung H.Y., Lee C.K., Joo S., Kim S., Yun C., Hahm M.G., Ahn M.-H., Kim S.-W., Oh Y.S., Shin S.R. Significance of various sensing mechanisms for detecting local and atmospheric greenhouse gases: A review // Adv. Sens. Res. 2024. V. 3. P. 2300094. DOI: 10.1002/adsr.202300094.
2. Dong M., Zheng C., Miao S., Zhang Y., Du Q., Wang Y., Titte F.K. Development and measurements of a mid-infrared multi-gas sensor system for CO, CO₂, and CH₄ detection // Sensors. 2017. V. 17. P. 2221. DOI: 10.3390/s17102221.
3. Abina A., Puc U., Jazbinsek M., Zidansek A. Analytical gas sensing in the Terahertz spectral range // Micromachines. 2023. V. 14, N 11. P. 1987. DOI: 10.3390/mi14111987.
4. Vaks V.L., Anfert'ev V.A., Balakirev V.Yu., Basov S.A., Domracheva E.G., Illyuk A.V., Kupriyanov P.V., Pripolzin S.I., Chernyaeva M.B. Spektroskopiya vysokogo razresheniya teragertsevogo chastotnogo diapazona dlya analiticheskikh prilojenii // Uspekhi fiz. nauk. 2020. V. 190, N 7. P. 763–776. DOI: 10.3367/UFNr.2019.07.038613.
5. Tret'yakov M.Yu. Vysokotochnaya rezonatornaya spektroskopiya atmosfernykh gazov v millimetrovom i submillimetrovom diapazonakh dlin voln. Nijnii Novgorod: IPF RAN, 2016. 320 p.
6. Rice T.E., Chowdhury M.A.Z., Mansha M.W., Hella M.M., Wilke I., Oehlschlaeger M.A. Halogenated hydrocarbon gas sensing by rotational absorption spectroscopy in the 220–330 GHz frequency range // Appl. Phys. B. 2021. V. 127. P. 123. DOI: 10.1007/s00340-021-07667-w.
7. Theuer M., Harsha S.S., Molter D., Torosyan G., Beigang R. Terahertz Terahertz time-domain spectroscopy of gases, liquids, and solids // Chem. Phys. Chem. 2011. V. 12. P. 2695–2705. DOI: 10.1002/cphc.201100158.
8. Hsieh Y.-D., Nakamura S., Abdelsalam D., Minamikawa T., Mizutani Y., Yamamoto H., Iwata T., Hindle F., Yasui T. Dynamic terahertz spectroscopy of gas molecules mixed with unwanted aerosol under atmospheric pressure using fibre-based asynchronous-optical-sampling terahertz time-domain spectroscopy // Sci. Rep. 2016. V. 6. P. 28114. DOI: 10.1038/srep28114.
9. Osseiran N., Bichon J., Eliet S., Peretti R. Can the THz-TDS detect trace gases? // 49th International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz), Perth, Australia, 2024. P. 1–2. DOI: 10.1109/IRMMW-THz60956.2024.10734282.
10. Dubrovkin J. Mathematical Processing of Spectral Data in Analytical Chemistry: A Guide to Error Analysis. Cambridge: Scholars Publishing, 2018. 380 p.
11. Debus B., Parastar H., Harrington P., Kirsanov D. Deep learning in analytical chemistry // Trends Analytical Chem. 2021. V. 145. P. 116459. DOI: 10.1016/j.trac.2021.116459.
12. Mishra P., Passos D., Roger J.-M., Marini F., Biancolillo A., Rutledge D., Gowen A., Amigo J., Jansen J. Deep learning for near-infrared spectral data modelling: Hypes and benefits // Trends Anal. Chem. 2022. V. 157. P. 116804. DOI: 10.1016/j.trac.2022.116804.
13. Wietzke S., Jansen C., Reuter M., Jung T., Kraft D., Chatterjee S., Fischer B.M., Koch M. Terahertz spectroscopy on polymers: A review of morphological studies // J. Mol. Struct. 2011. V. 1006. P. 41–51. DOI: 10.1016/j.molstruc.2011.07.036.
14. Nazarov M.M., Shkurinov A.P., Kuleshov E.A., Tuchin V.V. Teragertsovaya impul'snaya spektroskopiya biologicheskikh tkanei // Kvant. elektron. 2008. V. 38. P. 647–654. DOI: 10.1070/QE2008v038n07ABEH013851.
15. Gordon I., Rothman L., Hargreaves R., Hashemi R., Karlovets E., Skinner F., Conway E.K., Hill C., Kochanov R.V., Tan Y., Wcisło P., Finenko A.A., Nelson K., Bernath P.F., Birk M., Boudon V., Campargue A., Chance K.V., Coustenis A., Drouin B.J., Flaud J.-M., Gamache R.R., Hodges J.T., Jacquemart D., Mlawer E.J., Nikitin A.V., Perevalov V.I., Rotger M., Tennyson J., Toon G.C., Tran H., Tyuterev V.G., Adkins E.M., Baker A., Barbe A., Canè E., Császár A.G., Dudaryonok A., Egorov O., Fleisher A.J., Fleurbaey H., Foltynowicz A., Furtenbacher T., Harrison J.J., Hartmann J.M., Horneman V.-M., Huang X., Karman T., Karns J., Kassi S., Kleiner I., Kofman V., Kwabia-Tchana F., Lavrentieva N.N., Lee T.J., Long D.A., Lukashevskaya A.A., Lyulin O.M., Makhnev V.Yu., Matt W., Massie S.T., Melosso M., Mikhai­lenko S.N., Mondelain D., Müller H.S.P., Naumenko O.V., Perrin A., Polyansky O.L., Raddaoui E., Raston P.L., Reed Z.D., Rey M., Richard C., Tóbiás R., Sadiek I., Schwenke D.W., Starikova E., Sung K., Tamassia F., Tashkun S.A., Vander Auwera J., Vasilenko I.A., Vigasin A.A., Villanueva G.L., Vispoel B., Wagner G., Yachmenev A., Yurchenko S. The HITRAN2020 molecular spectroscopic database // J. Quant. Spectrosc. Radiat. Transfer. 2022. V. 277. P. 1–82. DOI: 10.1016/j.jqsrt.2021.107949.
16. Mikhailenko S.N., Babikov Yu.L., Golovko V.F. Informatsionno-vychislitel'naya sistema «Spektroskopiya atmosfernykh gazov». Struktura i osnovnye funktsii // Optika atmosf. i okeana. 2005. V. 18, N 9. P. 765–776.
17. He K., Zhang X., Ren S., Sun J. Deep residual learning for image recognition // Proc. IEEE. 2016. P. 770–778. DOI: 10.1109/CVPR.2016.90.
18. Kiranyaz S., Ince T., Abdeljaber O., Avci O., Gabbouj M. 1-D convolutional neural networks for signal processing applications // Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).  Brighton, UK, 2019.  P. 8360–8364. DOI: 10.1109/ICASSP.2019.8682194.
19. He F., Liu T., Tao D. Why resnet works? Residuals generalize // IEEE Trans. Neural Netw. Learn. Syst. 2020. V. 31, N 12. P. 5349–5362. DOI: 10.1109/TNNLS.2020.2966319.
20. Eliet S., Cuisset A., Guinet M., Hindle F., Mouret G., Bocquet R., Demaison J. Rotational spectrum of formaldehyde reinvestigated using a photomixing THz synthesizer // J. Mol. Spectrosc. 2012. V. 279. P. 12–15. DOI: 10.1016/j.jms.2012.07.010.